1125 is a great result. Gnubg contact benchmarks at about 1122, so your net is pretty much as good as gnubg. 1 point of contact-benchmark error is about 156.5
micropoints/game advantage.
We’ve managed to get as low as 1073 with the standard 250 inputs, but it took a 512 hidden nodes to do it, which would kill performance if we used it in the
production gnubg.
Are you prepared to discuss your extra inputs?
n
Ian
From:
bug-gnubg-bounces+address@hidden [mailto:bug-gnubg-bounces+ian.shaw=address@hidden]
On Behalf Of Øystein Schønning-Johansen
Sent: 12 February 2012 16:39
To: Mark Higgins
Cc: address@hidden
Subject: Re: [Bug-gnubg] "Joseph-ID" in benchmark db
2012/2/12 Mark Higgins <address@hidden>
My best player (TD trained, race & contact networks, a couple extra inputs beyond the standard Tesauro ones) has an average error of 0.0164ppg/move in the contact set, so not surprisingly worse than GNUbg (I assume 1125 means 0.01125ppg/move?).
No, I mean total error over the file.
1125 is the total error over all these 107485 positions. That makes an average error of 0.010466577 Equity points pr positon in the
contact.bm file.